PPP6R3-mediated dephosphorylation regulates mRNA translation during spermatogonial differentiation

Abstract

Protein dephosphorylation mediated by phosphatases regulates spermatogenesis. However, which proteins are dephosphorylated and how they regulate spermatogenesis are largely unknown. Here, we show that germline-specific deletion of protein phosphatase 6 regulatory subunit 3 (PPP6R3), which determines substrate specificity of protein phosphatase 6 (PP6), causes abnormal spermatogonial differentiation and male infertility, accompanied by translation inhibition. PPP6R3 directly interacts with EIF3C and EIF4G1 in KIT⁺ spermatogonia. Decreased levels of non-phosphorylated EIF3C and EIF4G1 after PPP6R3 deletion attenuate their enrichment for mRNAs associated with spermatogonial differentiation, and increased phosphorylation levels promote their degradation. Specifically, the phosphorylation status both of EIF3C^S39 and EIF4G1^S1217 are significantly up-regulated in mutant mice. Overexpression of EIF3C^S39A and EIF4G1^S1217A mutants in Ppp6r3-cKO spermatogonial progenitor cells compensates for the deficiency of differentiation potential by upregulating translation rates of differentiation-associated mRNAs. Our findings demonstrate EIF3C and EIF4G1, as specific substrates of PPP6R3/PP6 holoenzyme, are required for translation activation during spermatogonial differentiation.

A study using germ cell-specific knockout mice and a spermatogonial differentiation model reveals the relationship and role of PPP6R3-mediated dephosphorylation of translation initiation factors in spermatogonial differentiation.

Introduction

There are two pivotal transitions in the highly orchestrated and complex mitosis phase of mammalian spermatogenesis: spermatogonial differentiation and meiotic initiation. The former means the exit of spermatogonia from the stem cell pool, while the latter marks the end of mitosis and the initiation of meiosis¹. Spermatogonial stem cells (SSCs) lay the foundation for the sustained mammalian spermatogenesis. SSCs can not only self-renew to produce daughter stem cells, but also produce spermatogonial progenitor cells (SPCs), which will undergo spermatogonial differentiation and meiotic initiation in response to retinoic acid (RA) stimulation^2–4. In the absence of RA, RA receptors (RARs) heterodimerize with retinoid X receptors (RXRs) and bind to the RA response element^5,6, which futher maintains a dense and inhibited chromatin by interacting with corepressors such as nuclear receptor corepressor (NCoR) and silencing mediator for RA and thyroid hormone receptors (SMRT), ultimately leading to the silencing of target gene transcription⁷. In contrast, the binding of RA to RARs changes the conformation of RARs, causing RAR/RXR to release inhibitory protein complexes and recruit coactivators including histone acetylase, methyltransferase, and DNA-dependent ATPase⁸. At this time, chromatin accessibility facilitates the transcription of target genes such as Stra8 and Kit and promoting spermatogonial differentiation^9,10. Accurate gene transcription and protein expression are the basis and prerequisite for spermatogonial differentiation and meiotic initiation, which is regulated by various factors such as hormones^11,12, growth factors¹³, and epigenetic modification^14–17.

Phosphorylation, as one of the most common and reversible post-translational modifications (PTMs), is involved in the regulation of cell proliferation, differentiation, apoptosis, metabolism, and tumorigenesis by altering protein conformation, localization, stability, and interaction with other biomolecules^18–20. Eukaryotic protein phosphatases can be divided into phosphoprotein phosphatase (PPP), Mg²⁺/Mn²⁺ dependent protein phosphatase (PPM), aspartate-based protein phosphatase, and phosphotyrosine phosphatase (PTP), according to substrate specificity, catalytic activity, as well as inhibitor sensitivity^21,22. More than 80% of the protein phosphatase activity in eukaryotic cells is regulated by PPP, which consists of PP1, PP2A, PP2B, PP4, PP5, PP6, and PP7^23,24. Studies have shown that both PP2A, PP4, and PP6 are involved in the regulation of meiosis^25–28. Among them, PP6 is composed of catalytic subunit (PPP6C), regulatory subunit (PPP6R, also named SAPS), and scaffolding subunit²⁹. Lei et al. have revealed that germline-specific deletion of PPP6C using Stra8-cre mice resulted in male infertility, and spermatocytes were blocked at the pachytene stage, accompanied by defects of double strand break (DSB) repair and crossover formation³⁰. In addition, Sertoli cell-specific deletion of PPP6C by Amh-cre mice also led to male infertility due to the loss of mature spermatozoa mediated by hyperphosphorylation of β-CATENIN³¹. These findings demonstrated that PPP6C is essential for spermatogenesis and male fertility.

PPP6C activity, however, is modulated by PPP6R through restricting PP6 substrate specificity and determining the intracellular localization of PP6^32,33. There are three conserved PPP6R known between human, mouse, and rat species: PPP6R1, PPP6R2, and PPP6R3. Differences in their protein structure, expression patterns in cells or tissues, and substrate recruitment determine the variation or specificity of PPP6C activity and response to specific signals. Even though PPP6R1 and PPP6R3 are more closely related in sequence to one another than either is related to PPP6R2, there is apparent specificity in function. One study, for example, showed that knockdown of PPP6R1 but not PPP6R3 promotes the degradation of IκBϵ in response to TNFα stimulation³². IκBϵ is a likely candidate as a specific substrate for the PPP6R1/PP6 holoenzyme. Moreover, Yang et al. revealed PPP6R3 but not PPP6R1 or PPP6R2 is a negative regulator of AMPK, and a high-fat diet induces up-regulation of PPP6R3 and recruits PPP6C to inactivate phosphorylated AMPK, leading to metabolic disorders³⁴. Inhibition of PPP6R3 may be a potential therapeutic strategy to treat metabolic syndromes. These studies highlight that PPP6R plays critical roles in signal transduction and maintenance of metabolic homeostasis. However, there has been no report to evaluate whether PPP6R is involved in spermatogenesis.

In this study, we found that PPP6R3 had the highest protein abundance in all different types of germ cell populations compared to PPP6R1 and PPP6R2 based on previously reported data in mouse testes. Next, we generated germline-specific Ppp6r3 knockout mice to analyze its role in spermatogenesis. Our data demonstrated the essential role of PPP6R3 in spermatogonial differentiation and male fertility. Deletion of PPP6R3 inhibited the translation but not transcription of multiple known regulators of spermatogonial differentiation. Further, translation initiation factors EIF3C and EIF4G1 are specific substrates for the PPP6R3/PP6 holoenzyme in KIT⁺ spermatogonia. The increased phosphorylation levels of EIF3C^S39 and EIF4G1^S1217 after PPP6R3 deletion promoted their degradation, ultimately leading to the failure of translation during spermatogonial differentiation. Our findings reveal the essential role of PPP6R3 in spermatogonial differentiation, and elucidate the link between phosphatase-mediated dephosphorylation and translational activation during spermatogenesis.

Results

PPP6R3 is required for spermatogenesis

To explore the potential role of PPP6R in spermatogenesis, first we analyzed the levels of PPP6R1, PPP6R2, and PPP6R3 based on previously reported quantitative proteomics data for mouse testes³⁵ and found that, for all types of germ cells populations, PPP6R2 had the lowest level while PPP6R3 consistently had the highest levels (Fig. 1A). We also noted that the protein abundance of PPP6C remained stable (Fig. 1A). We then profiled PPP6R3 expression in multiple mouse tissues. PPP6R3 was highly expressed in wild-type testes, and its level was increased from post-natal day 8 (P8) compared to that of P6 (Fig. 1B, Fig. S1A), and it was localized in the cytoplasm (Fig. S1B, S1C). Further immunostaining and Western blotting results showed that the level of PPP6R3 in PLZF⁺ undifferentiated spermatogonia in P12 testes was similar to that in SOX9⁺ Sertoli cells, but significantly lower than that in KIT⁺ differentiating spermatogonia and SYCP3⁺ spermatocytes (Fig. 1C-1H). These results suggest possible functions of PPP6R3 in spermatogenesis, especially in spermatogonial differentiation and spermatocytes development.

Fig. 1. PPP6R3 is highly expressed in differentiating spermatogonia and early spermatocytes.

A The protein abundances of PPP6R1, PPP6R2, PPP6R3, and PPP6C in different types of germ cell populations based on previously reported quantitative proteomics data in mice testes. Aun, type A undifferentiated spermatogonia. eLL, early leptotene and leptotene. Z, zygotene. eP, early pachytene. mP, middle pachytene. IP, late pachytene. eD, early diplotene. ID, late diplotene. RS, round spermatid. B Western blotting of PPP6R3 in the testis of wild-type mice at indicated times. *p < 0.05, **p < 0.01. Data are presented as means ± SD. n = 3 biological replicates. C Immunostaining of PPP6R3 and KIT in the testis sections from P12 wild-type mice. Scale bar, 15 μm. D Immunostaining of PPP6R3 and SYCP3 in the testis sections from P12 wild-type mice. Scale bar, 15 μm. E Immunostaining of PPP6R3 and PLZF in the testis sections from P12 wild-type mice. Scale bar, 15 μm. F Immunostaining of PPP6R3 and SOX9 in the testis sections from P12 wild-type mice. Scale bar, 15 μm. G Quantitative results of fluorescence intensity of PPP6R3 in PLZF⁺ spermatogonia, KIT⁺ spermatogonia, early spermatocytes and SOX9⁺ Sertoli cells. ***p ^< 0.001. ns, not statistically significant. Data are presented as means ± SD. n = 20 technical replicates. H Western blotting of PPP6R3 in THY1⁺ or KIT⁺ spermatogonia, as well as spermatocytes purified by using a cell purification kit or fluorescence-activated cell sorting (FACS) from P12 wild-type mice. *p < 0.05, ***p < 0.001. Data are presented as means ± SD. n = 3 biological replicates. See also Fig. S1.

To test the potential role of PPP6R3 in spermatogenesis, we generated germline-specific Ppp6r3 knockout mice (referred to as Ppp6r3-cKO) by crossing Ppp6r3^flox/flox mice with Stra8-cre mice. Compared to the control, PPP6R3 protein level was significantly reduced in P6 Ppp6r3-cKO testes (Fig. S1D), accompanied by a specific deletion of PPP6R3 in CCND2-positive cells (a marker for type A1 spermatogonia) (Fig. S1E). Notably, this deletion did not affect PPP6R3 protein expression in undifferentiated spermatogonia marked by PLZF or THY1 (Fig. S1F, S1G). These findings suggest a germline-specific deletion of PPP6R3. Adult Ppp6r3-cKO male mice were infertile (Fig. S1H), and the testicular/body weight ratio of Pppp6r3-cKO males decreased significantly starting from two weeks of age (Fig. 2A, 2B). Hematoxylin staining of testes sections revealed an obvious loss of germ cells in adult Ppp6r3-cKO mice, with only a few spermatocytes scattered in seminiferous tubules, and no spermatozoa were observed in Ppp6r3-cKO epididymides (Fig. 2C). However, somatic cell (including Sertoli cell and Leydig cell)-specific Ppp6r3 knockout, achieved by crossing Ppp6r3^flox/flox mice with Sf1-cre mice (Fig. S1I, S1J), did not affect spermatogenesis or male fertility (Fig. S1K). These results demonstrate that PPP6R3 is required for male germ cell development and spermatogenesis. PPP6R3 is required for spermatogenesis by regulating the development of male germ cells rather than somatic cells.

Fig. 2. Germline-specific deletion of PPP6R3 causes male infertility and spermatogonial differentiation failure.

A Testis size comparison between adult wild-type and Ppp6r3-cKO mice. B The changes in testis: body weight ratio (mg/g) of the control and Ppp6r3-cKO mice aged from P7 to P56. ***p < 0.001. Data are presented as means ± SD. n = 3 biological replicates. C Hematoxylin staining of paraffin-embedded testes and epididymis sections from adult wild-type and Ppp6r3-cKO mice. D Hematoxylin staining of paraffin-embedded testes sections from wild-type and Ppp6r3-cKO mice at indicated times. Scale bar, 20 μm. E Immunostaining of MVH in testes sections from wild-type and Ppp6r3-cKO mice at indicated times. Scale bar, 20 μm. F Quantitative results of the number of MVH⁺ cells per tubule. *p < 0.05, **p < 0.01, ***p < 0.001. ns, not statistically significant. Data are presented as means ± SD. n = 3 biological replicates. G Immunostaining of PLZF and MVH in testes sections from P9 wild-type or Ppp6r3-cKO mice. Scale bar, 20 μm. ***p < 0.001. ns, not statistically significant. Data are presented as means ± SD. n = 3 biological replicates. H Immunostaining of PLZF, STRA8, and KIT in testes sections from P9 wild-type or Ppp6r3-cKO mice. Scale bar, 20 μm for left panel, 15 μm for right panel. I Quantitative results of the number of PLZF⁺&STRA8⁺ or KIT⁺ cells per tubule. ***p < 0.001. Data are presented as means ± SD. n = 3 biological replicates. See also Figure S1.

PPP6R3 promotes the differentiation of spermatogonia

To identify which spermatogenesis stage(s) are abnormal in Ppp6r3-cKO mice, we performed hematoxylin staining on testes sections from P6, P9, P14, and P21 mice. The testicular tubules of Ppp6r3-cKO mice were similar to controls at P6, but there was an obvious reduction in the number of germ cells by P9 (Fig. 2D). At P14 and P21, most of the germ cells in Ppp6r3-cKO testes were lost, and only a few spermatocyte-like cells were scattered inside the lumen of testicular tubules (Fig. 2D). This drastic reduction in the total number of germ cells in Ppp6r3-cKO testes was also evident upon immunostaining against MVH (also known as DDX4 or VASA), a marker of germ cells (Figs. 2E, 2F). Co-immunostaining against MVH and PLZF (also called ZBTB16), a well-known transcription factor required for the self-renewal of spermatogonia, showed that the number of PLZF⁺MVH⁺ germ cells (referring to undifferentiated spermatogonia) in testes of Ppp6r3-cKO mice did not change significantly after PPP6R3 deletion, whereas the number of PLZF⁺MVH⁺ germ cells (referring to differentiating spermatogonia) was significantly decreased (Fig. 2G), indicating that loss of PPP6R3 function apparently affect spermatogonial differentiation.

We next examined whether PPP6R3 is essential for spermatogonial differentiation. Co-immunostaining against PLZF and STRA8, a marker for differentiating spermatogonia and preleptotene spermatocytes, showed that PLZF⁺STRA8⁺ spermatogonia were rarely observed in P9 Ppp6r3-cKO testes (Fig. 2H, 2I). Also, the number of KIT⁺ differentiating spermatogonia (represent type A1-A4 spermatogonia) was significantly reduced (Fig. 2H, 2I). Further immunostaining against PLZF showed that the number of PLZF⁺ spermatogonia in testes of P28 and P42 Ppp6r3-cKO mice increased abnormally compared to wild-type mice (Fig. S1L, S1M), indicating the accumulation of undifferentiated spermatogonia. These results indicate that conditional knockout of PPP6R3 from germ cells blocks spermatogonial differentiation.

Deletion of PPP6R3 impairs the transition of differentiating spermatogonia from mitosis to meiosis

Given that spermatogonial differentiation is prerequisite for meiosis initiation, we assumed that meiosis would be disrupted in the Ppp6r3-cKO mice. The number of preleptotene spermatocytes (STRA8⁺SYCP3⁺ cells or STRA8⁺γH2AX⁺ cells) in P10 Ppp6r3-cKO testes was significantly decreased compared to that of control (Fig. S2A–S2D), Further, these few preleptotene spermatocytes can only develop into zygotene-like spermatocytes, and no pachytene or diplotene spermatocytes were observed in testes of 3-week-old Ppp6r3-cKO mice (Fig. S2E–S2H). Collectively, these results demonstrate that deletion of PPP6R3 impairs the meiotic initiation of differentiating spermatogonia.

PPP6R3 is essential for the translation but not transcription of multiple spermatogonial differentiation-related mRNAs

We next explored how PPP6R3 regulates spermatogonial differentiation by conducting transcriptomics and proteomics profiling of testis samples from P9 wild-type and Ppp6r3-cKO mice. The level of Ppp6r3 mRNA in Ppp6r3-cKO testes was significantly reduced compared to controls (Fig. 3A); there were no differences in the expression levels for multiple known spermatogonial maintenance genes (Plzf, Oct4, Id4, Gfrα1, Etv5, Nanos2, and Nanos3) or for a set of spermatogonial differentiation-related genes (Kit, Dmrt1, Ccnd2, Sohlh1, and Sohlh2), although the level of Stra8 mRNA was significantly elevated in Ppp6r3-cKO testes. qPCR analysis targeting all of the aforementioned mRNA molecules showed consistent trends with the transcriptomics data (Fig. 3C), suggesting that the deletion of PPP6R3 did not affect (at least not inhibit) the transcription of spermatogonial differentiation-related genes. However, the proteomics analysis showed that the levels of STRA8, CCND2, KIT, DMRT1, SOHLH1, and SOHLH2 were dramatically down-regulated in Ppp6r3-cKO testes (Figs. 3D, 3F, 3G). These results indicate that PPP6R3 may function in some translation-related process during spermatogonial differentiation.

Fig. 3. PPP6R3 promotes the translation of a set of mRNAs associated with spermatogonial differentiation.

A Volcano plot of differentially expressed mRNAs in testis-based transcriptomics after PPP6R3 deletion. n = 3 biological replicates. B Enriched gene ontology (GO) terms of the down-regulated mRNAs in the testis after PPP6R3 deletion. C qRT-PCR analysis of indicated mRNAs in testes from P9 wild-type and Ppp6r3-cKO mice. **p < 0.01. ns, not statistically significant. Data are presented as means ± SD. n = 3 biological replicates. D Volcano plot of differentially expressed proteins in testis-based proteomics after PPP6R3 deletion. n = 3 biological replicates. E Enriched GO terms of the down-regulated proteins in the testis after PPP6R3 deletion. F Western blotting of indicated proteins in testes of P9 wild-type or Ppp6r3-cKO mice. G Quantitative results of the protein expression in (F). *p < 0.05, ***p < 0.001. ns, not statistically significant. Data are presented as means ± SD. n = 3 biological replicates.

We also performed gene ontology (GO) term enrichment analyses for the 620 down-regulated transcripts from the transcriptomics data and the 70 down-regulated proteins from the proteomics data (Fig. 3A, 3D). The down-regulated transcripts showed enrichment for terms including “meiosis I” (Brca2, Stag3, Rec8, Meiosin, Meioc, Dmrtc2, Gal3st1, and M1ap), “synaptonemal complex assembly” (Sycp1, Sycp3, Syce1, and Syce3), and “homologous recombination” (Dmc1, Msh5, Mei1, and Spo11) (Fig. 3B). Moreover, the levels of TDRKH, STAG3, M1AP, and TEX12 proteins associated with “meiosis I”, “meiotic cell cycle”, and “meiotic chromosome segregation” were significantly reduced in the testis of Ppp6r3-cKO mice (Fig. 3E). These results show that abnormal spermatogonial differentiation after PPP6R3 deletion results in the repression of the transcription and protein expression of meiotic genes.

Knockout of Ppp6r3 from differentiating spermatogonial progenitor cells blocks translation of differentiation-related mRNAs

To further determine that the failure of spermatogonial differentiation after PPP6R3 deletion is related to translation repression, we established a differentiation system in vitro for spermatogonial progenitor cells (SPCs). First, we isolated and purified SPCs from the testis of a single wild-type mouse and a single Ppp6r3-cKO mouse using anti-mouse CD90.2 magnetic particles, and cultured them for 5 months (Fig. S3A). All of the examined SPCs express MVH and markers of undifferentiated spermatogonia (PLZF, OCT4, GFRα1, ETV5, and ID4) (Fig. S3B-S3D), and deletion of PPP6R3 did not affect the viability and proliferative potential of SPCs (Fig. S3E-S3F). We then induced wild-type SPCs to differentiate using RA. qPCR analysis showed that the levels of spermatogonial maintenance-related mRNAs (Plzf, Gfrα1, and Oct4) were gradually decreased with the extension of RA induction time, while the levels spermatogonial differentiation-related mRNAs (Stra8, Kit, Dmrt1, and Ccnd2) were increased significantly after 12 h of differentiation (Fig. S3G, S3H). Moreover, co-immunostaining against PLZF, STRA8, and KIT showed that about 95% of the undifferentiated SPCs (PLZF⁺STRA8^-KIT^- cells) entered the differentiation stage (PLZF⁺STRA8⁺KIT^- cells) after 6 h-treatment, then these cells effectively developed into early differentiating spermatogonia (PLZF^-STRA8⁺KIT⁺ cells represent type A1-A4 spermatogonia) after 12 h of differentiation (Fig. 4A, 4B). No late differentiating spermatogonia (PLZF^-STRA8^-KIT⁺ cells represent type intermediate to type B spermatogonia) and preleptotene spermatocytes (PLZF^-STRA8⁺KIT^-) were observed, even after 24 h of differentiation (Fig. 4A, 4B). These results suggest that this system can effectively induce wild-type SPCs to differentiate.

Fig. 4. The translation rates of spermatogonial differentiation-related mRNAs decreased during Ppp6r3-cKO SPCs differentiation in vitro.

A Immunostaining of PLZF, STRA8, and KIT during wild-type SPCs differentiation in vitro at indicated times. Scale bar, 15 μm. B Quantitative results of the proportion of different types of spermatogonia during wild-type SPCs differentiation in vitro. Data are presented as means ± SD. n = 3 biological replicates. C Immunostaining of PLZF, STRA8, and KIT during Ppp6r3-cKO SPCs differentiation in vitro at indicated times. Scale bar, 15 μm. D Quantitative results of the proportion of different types of spermatogonia during Ppp6r3-cKO SPCs differentiation in vitro. Data are presented as means ± SD. n = 3 biological replicates. E Western blotting of PLZF, STRA8, KIT, and PPP6R3 at indicated differentiation times of SPCs from wild-type or Ppp6r3-cKO mice. n = 3 biological replicates. F The translation rates of Plzf, Oct4, and Gfrα1 mRNAs at indicated differentiation times of SPCs from wild-type or Ppp6r3-cKO mice. *p < 0.05, **p < 0.01, ***p < 0.001. Data are presented as means ± SD. n = 3 biological replicates. G The translation rates of Stra8, Kit, Dmrt1, and Ccnd2 mRNAs at indicated differentiation times of SPCs from wild-type or Ppp6r3-cKO mice. *p < 0.05, **p < 0.01, ***p < 0.001. ns, not statistically significant. Data are presented as means ± SD. n = 3 biological replicates. See also Fig. S3.

However, the majority of Ppp6r3-cKO SPCs were PLZF⁺STRA8⁺KIT^- after 12-24 h of differentiation, and fewer than 10% of Ppp6r3-cKO SPCs developed into early differentiating spermatogonia, which is similar to the abnormal spermatogonial differentiation in male Ppp6r3-cKO mice (Fig. 4C, 4D). In addition, after 6 h of differentiation, both the protein level and positive rate of PPP6R3 in Ppp6r3-cKO SPCs were significantly decreased, which was consistent with the increased activity of Stra8-cre (Fig. 4E, Figure S3I-S3K). And after 12 h of differentiation, the expression of PPP6R3 in Ppp6r3-cKO SPCs was nearly undetectable by Western blotting, although immunostaining revealed a small population of cells displaying weak PPP6R3 positivity (Fig. 4E, Figure S3L). These results indicate that this cell model mimics the abnormal spermatogonial differentiation phenotype caused by PPP6R3 deletion.

We then used this cell model and a previously reported “Targeted Profiling of RNA Translation”^36,37 to monitor changes in the translation rates both of spermatogonial maintenance and differentiation-related mRNAs at 0, 6, and 12 h after induction of wild-type and Ppp6r3-cKO SPCs differentiation. The translation rates of spermatogonial maintenance-related mRNAs (such as Oct4 and Gfrα1 mRNAs) in wild-type SPCs were significantly lower than those in Ppp6r3-cKO SPCs after 6 h of differentiation, and the translation rate of Plzf mRNA was statistically significant after 12 h of differentiation (Fig. 4F). For mRNAs associated with spermatogonial differentiation (including Stra8, Kit, Dmrt1, and Ccnd2 mRNAs), their translation rates in wild-type SPCs were significantly higher than those in Ppp6r3-cKO SPCs after 6 h or 12 h of differentiation (Fig. 4G). These results suggest that PPP6R3 is required for the translation of spermatogonial differentiation-related mRNAs.

EIF3C and EIF4G1 directly interact with PPP6R3 and are the specific substrates for PP6 in KIT⁺ differentiating spermatogonia

It is well-known that PPP6R regulates PPP6C’s catalytic activity by restricting PP6 substrate specificity. To identify which molecules act as potential substrates for PP6 during spermatogonial differentiation, we purified KIT⁺ spermatogonia from the testis of P9 wild-type mice (Fig. S4A) and then performed immunoprecipitation mass spectrometry (IP-MS) using a PPP6R3 antibody (Fig. S4B). In addition to molecules previously known to interact with PPP6R3, such as PP6 catalytic subunit (PPP6C), two regulatory subunits (PPP6R1 and PPP6R2), and two scaffolding subunits (ANKRD28 and ANKRD44), the PPP6R3 antibody also pulled-down multiple translation initiation factors, including EIF3B, EIF3C, EIF3D, EIF3F, EIF3G, EIF3L, EIF3M, EIF4E, and EIF4G1 (Fig. 5A). GO term and KEGG pathway analysis of these molecules with potential interactions with PPP6R3 showed enrichment for functional annotations related to ribosome structure assembly, translation regulator activity, and translation initiation factor binding (Fig. S4C, S4D). Co-IP experiments revealed that PPP6R3 directly interacts with EIF3C and EIF4G1 in an RNA-independent manner, as these interactions remained stable after RNase A treatment. In contrast, the observed associations with EIF3B, EIF3F, EIF3M, and EIF4E were abolished by RNase A digestion, indicating that these interactions are RNA-dependent and likely mediated by shared RNA binding (Fig. 5B).

Fig. 5. EIF3C and EIF4G1 directly interacts with PPP6R3 in KIT⁺ spermatogonia.

A Volcano plot of significantly enriched proteins by PPP6R3 antibody in KIT⁺ spermatogonia. B Co-IP analysis of interactions between PPP6R3 and multiple translation initiation factors in KIT⁺ spermatogonia. C Western blotting of phosphorylated and non-phosphorylated translation initiation factors in KIT⁺ spermatogonia purified from P10 wild-type or Ppp6r3-cKO mice. D Quantitative results of the levels of phosphorylated and non-phosphorylated translation initiation factors. ***p < 0.001. ns, not statistically significant. Data are presented as means ± SD. n = 3 biological replicates. See also Fig. S4.

We next examined the levels of non-phosphorylated and phosphorylated translation initiation factors in KIT⁺ spermatogonia using immunoblotting. Compared with the control group, the non-phosphorylation levels of EIF3C and EIF4G1, which directly interact with PPP6R3, were significantly reduced, and their phosphorylation levels were significantly increased. However, there were no significant changes in the non-phosphorylation and phosphorylation levels of translation initiation factors (EIF3B, EIF3F, EIF3M, and EIF4E) (Fig. 5C, 5D). These results indicate that EIF3C and EIF4G1 are specific substrates for PP6 in KIT⁺ spermatogonia.

We further clarified the relationship between the reduced translation rates after PPP6R3 deletion and EIF3C and EIF4G1. The results of RIP-qPCR showed that both EIF3C and EIF4G1 significantly enriched spermatogonial maintenance and differentiation-related mRNAs in wild-type mice testes (Fig. S5A-S5D). The deletion of PPP6R3 did not affect the enrichment of spermatogonial maintenance-related mRNAs by EIF3C and EIF4G1 (Figure S5C, S5D), but their enrichment for spermatogonial differentiation-related mRNAs was significantly weakened with the decrease of their non-phosphorylation levels (Figure S5A, S5B). Moreover, the increased phosphorylation levels of EIF3C and EIF4G1 led to decreased stability and promoted their degradation (Fig. S6A, S6B). Overall, our data demonstrate that PPP6R3 deletion promotes phosphorylation of EIF3C and EIF4G1, which in turn promotes their degradation. On the other hand, the reductions of EIF3C and EIF4G1 impair their binding to mRNAs associated with spermatogonial differentiation, which may be a direct cause of the low translation rate.

Hyperphosphorylation of EIF3C^S39 and EIF4G1^S1217 after PPP6R3 deletion is responsible for the decreased translation rates of differentiation-related mRNAs

Next, we screened the sites of altered phosphorylation in EIF3C and EIF4G1 after PPP6R3 deletion through testis-based phosphoproteomics from P9 wild-type and Ppp6r3-cKO mice. Phosphoproteomics data showed that phosphorylation was significantly down-regulated at 320 sites and up-regulated at 64 sites in testes of Ppp6r3-cKO mice compared with wild-type mice (Fig. 6A). Subsequently, GO term enrichment analysis was performed on the proteins corresponding to 320 significantly down-regulated phosphorylation sites and 64 significantly up-regulated phosphorylation sites. The proteins with significant down-regulation of phosphorylation were involved in the processes of “meiosis I”, “germ cell development”, “regulation of stem cell differentiation”, and “spermatid differentiation” (Fig. S6C). However, the proteins with significant up-regulation of phosphorylation (including EIF3C and EIF4G1) were involved in multiple translation-related processes, such as “regulation of translation in response to stress” and “translational initiation” (Fig. 6A, 6B).

Fig. 6. Phosphorylation of EIF3C and EIF4G1 reduces the differentiation efficiency of Ppp6r3-cKO SPCs by downregulating translation rates of differentiation-related mRNAs.

A Volcano plot of differentially phosphorylated proteins in testis-based phosphoproteomics after PPP6R3 deletion. n = 3 biological replicates. B Enriched gene ontology (GO) terms of proteins with significant up-regulation of phosphorylation in the testis after PPP6R3 deletion. C The detailed information of significant up-regulation sites for phosphorylation of EIF3C and EIF4G1 after PPP6R3 deletion. D Western blotting of the phosphorylation levels of EIF3C at S39 and of EIF4G1 at S1217 in testes of wild-type and Ppp6r3-cKO mice. **p < 0.01, ***p < 0.001. Data are presented as means ± SD. n = 3 biological replicates. E Immunostaining of PLZF, STRA8, and KIT after 12 h differentiation of Ppp6r3-cKO SPCs with overexpression of indicated mutant variants. Scale bar, 15 μm. F Quantitative results of the proportion of different types of spermatogonia after 12 h differentiation of Ppp6r3-cKO SPCs with overexpression of indicated mutant variants. Data are presented as means ± SD. n = 3 independent transfection experiments on the same cell line (3 technical replicates). G The translation rates of Stra8, Kit, Dmrt1, and Ccnd2 mRNAs after 12 h differentiation of Ppp6r3-cKO SPCs with overexpression of indicated mutant variants. *p < 0.05, **p < 0.01, ***p < 0.001. Data are presented as means ± SD. n = 3 independent transfection experiments on the same cell line (3 technical replicates). See also Fig. S6.

We then focused on the changes in the modification levels of phosphorylation sites in EIF3C and EIF4G1 after PPP6R3 deletion, and found that the phosphorylation levels of EIF3C at S39 and of EIF4G1 at S1217 were significantly up-regulated in testes of Ppp6r3-cKO mice, compared to those of wild-type mice (Figs. 6C, 6D). However, the phosphorylation status of serine and threonine at other sites in EIF3C and EIF4G1 and the phosphorylation levels of other translation initiation factors (EIF3B, EIF3F, EIF3M, and EIF4E) did not change significantly (Fig. 6A, S6D). These results suggest that EIF3C^S39 and EIF4G1^S1217 may be the targets of PP6 in regulating spermatogonial differentiation.

To test their functions, mutant variants of EIF3C and EIF4G1 (i.e., EIF3C^S39A, EIF4G1^S1217A) were overexpressed in Ppp6r3-cKO SPCs (Fig. S6E). Immunostaining showed that the proportion PLZF⁺STRA8⁺KIT^- cells in Ppp6r3-cKO SPCs overexpressing EIF3C^S39A, EIF4G1^S1217A, and EIF3C^S39A&EIF4G1^S1217A decreased from 83.31% to 55.75%, 66.35%, and 37.94%, respectively, after 12 h of differentiation (Fig. 6E, 6F). Meanwhile, the proportion of early differentiating spermatogonia increased significantly from 16.04% to 43.45%, 32.70%, and 61.93%, respectively (Fig. 6E, 6F). Western blotting results also indicated that overexpression of EIF3C^S39A, EIF4G1^S1217A, and EIF3C^S39A&EIF4G1^S1217A in Ppp6r3-cKO SPCs promoted the expression of spermatogonial differentiation-related proteins, and down-regulated the levels of spermatogonial maintenance-related proteins (Fig. S6F, S6G). These results indicate that overexpression of EIF3C and EIF4G1 mutants can reactivate the differentiation potential of Ppp6r3-cKO SPCs, and that the dephosphorylation of these two sites is critical for spermatogonial differentiation.

Finally, we investigated whether the recovery of spermatogonial differentiation efficiency after overexpression of EIF3C and EIF4G1 mutant variants was accompanied by changes in translation rate. As shown in Figure S6H, the translation rates of mRNAs associated with spermatogonial maintenance (Plzf, Oct4, and Gfrα1) were dramatically reduced during Ppp6r3-cKO SPCs overexpressing EIF3C and EIF4G1 mutant variants differentiation, compared to those of control. However, the translation rates of mRNAs associated with spermatogonial differentiation (Stra8, Kit, Dmrt1, and Ccnd2) were significantly increased during Ppp6r3-cKO SPCs overexpressing EIF3C and EIF4G1 mutant variants differentiation (Fig. 6G). These results suggest that PP6 promotes the translation of mRNAs associated with spermatogonial differentiation by targeting dephosphorylation of EIF3C^S39 and EIF4G1^S1217, which is essential for spermatogenesis and male fertility.

Discussion

PTMs play a crucial role in spermatogenesis, influencing processes such as sperm maturation³⁸, sperm motility^39,40, and sperm capacitation⁴¹. PP6 belongs to the PP2A-like subfamily, with the PPP6C catalytic subunit specifically facilitating the dephosphorylation of targeted protein substrates in distinct cells or physiological contexts. Previous research has indicated that protein dephosphorylation mediated by phosphatase catalytic subunit is vital for gametogenesis. For instance, the absence of PPP6C leads to incomplete development of spermatocytes into the diploid stage within the testes³⁰, alongside complications in folliculogenesis and instances of premature ovarian failure²⁸. In addition, the phosphatases PPP2CA and PPP4C are essential for spermatogenesis; their deletion halts meiosis at the diploid stage and results in oligoasthenoteratospermia, characterized by abnormalities in sperm tail structure, reduced sperm count, and impaired motility^26,27. Hu et al. noted that PP2A-Aalpha, a scaffold subunit of PP2A, is required for oocyte meiotic maturation but not for folliculogenesis²⁵. Here, we investigated the functional role of PPP6R3 in spermatogenesis through a multi-faceted approach, including germline-specific Ppp6r3 gene knockout mice, integrative multi-omics analysis, and a validated in vitro spermatogonial differentiation system, which serves as an ideal cellular model for exploring the mechanisms underlying spermatogonial differentiation defects caused by gene mutations or deletions. Our findings demonstrate that PPP6R3 is indispensable for the proper differentiation of SPCs, with its loss resulting in arrested differentiation and impaired translational reprogramming. Our study is the first to uncover the potential role of PPP6R in spermatogenesis, highlighting that PPP6R3 serves as a significant regulator of spermatogonial differentiation.

Spermatogonial differentiation marks the irreversible initiation of spermatogenesis. Key factors driving this process include the RA signaling pathway, NOTCH, bone morphogenetic protein 4 (BMP4), KIT, and stem cell factor (SCF), the ligand of KIT^42–44. Previous studies have confirmed that RA play a role in triggering spermatogonial differentiation by activating the transcription of Stra8 and Kit^45,46. KIT is considered a molecular marker of entry into the differentiation pathway⁴⁷, and is primarily expressed in early differentiating spermatogonia (types A1 to A4)^48–50. These early spermatogonia are then transformed into type B spermatogonia under the stimulation of SCF^48,51. The expression of KIT is regulated by several positive factors, including RA, SOHLH1, SOHLH2, BMP4, and SOHL2^13,46,52,53, as well as negative regulators such as GDNF, PLZF, and COL1A1^54–56. Notably, STRA8 protein in testes of Ppp6r3-cKO mice was down-regulated relative to wild-type mice, but its transcription was paradoxically elevated. We hypothesize that this discrepancy might arise from stress induced by Ppp6r3 deletion. In other words, the Stra8-cre-driven conditional knockout of Ppp6r3 in germ cells disrupted spermatogonial differentiation. To compensate for impaired differentiation, up-regulated transcriptional compensation of Stra8 mRNA, which plays a decisive role in initiating spermatogonial differentiation, was observed. However, Ppp6r3 deletion unexpectedly reduced the non-phosphorylated forms of EIF3C and EIF4G1, which are essential for translational regulation. The loss of these non-phosphorylated isoforms diminished their binding affinity for Stra8 mRNA, impairing ribosomal recruitment and triggering post-transcriptional sequestration of Stra8 mRNA. Consequently, despite enhanced Stra8 mRNA abundance, its translational output was suppressed, ultimately blocking the progression of spermatogonial differentiation.

It is also worth emphasizing that although KIT is widely used as a marker for A-type differentiating spermatogonia, its expression extends to later stages, including intermediate/B late type differentiating spermatogonia and early meiotic stages (preleptotene/leptotene). In this study, although the selection of 9-day-old mice can reflect the specific differentiation stage to a certain extent, it is undeniable that preleptotene spermatocytes have appeared at this time. This implies that some of the potential interacting molecules of PPP6R3 obtained by KIT⁺ spermatogonia IP-MS may play roles in both differentiation and meiosis initiation, rather than specifically functioning. This is a limitation in this study, and it is necessary to screen more differentiating spermatogonia-specific molecular markers for different developmental stages in the future to overcome limitations in cell type specificity of KIT⁺ markers.

Previous research has highlighted the importance of kinase signaling in regulating spermatogonial proliferation and differentiation. Exogenous RA induces the expression of the receptor tyrosine kinase KIT, which upon binding its ligand, activates the PI3K/AKT signaling pathway^57–60. This cascade further stimulates mTORC1 through increased phosphorylation of mTOR⁵³. Among the key targets of mTOR phosphorylation are EIF4EBP1, which facilitates cap-dependent mRNA translation, and ribosomal protein S6-kinase (S6K), which directly modifies ribosomes to enhance their synthetic activity^46,61,62. EIF4EBP1 negatively regulates the formation of the EIF4F complex by competing with EIF4G for binding to EIF4E, thereby suppressing cap-dependent translation initiation⁶³. Phosphorylation of EIF4EBP1 induces its dissociation from EIF4E, and then increases the translational efficiency of key mRNAs involved in spermatogonial differentiation, such as Kit, Sohlh1, and Sohlh2^53,63,64. Zheng et al. demonstrated that amino acid starvation induces autophagy in goat fetal fibroblasts and concurrently elevates phosphorylation levels of Ser216 on EIF5B, Ser39 on EIF3C, and Ser1194 on EIF4G1⁶⁵. However, the molecular mechanism by which these phosphorylation events regulate autophagy remains undefined.

Several studies have shown multiple serine phosphorylation sites on EIF4G1 mediated by different signaling pathways seems to be involved in regulating different biological processes. For example, the phosphorylation of Ser1232 on EIF4G1 is closely related to translation inhibition during cell mitosis^66,67. In human somatic cells, the phosphorylation of Ser1232 on EIF4G1 enhances its interaction with EIF4A and inhibits the RNA binding ability of the EIF4A helicase complex, thereby suppressing translation⁶⁶. However, the phosphorylation of Ser1232 on EIF4G1 in tumor cells inhibits the translation of hypoxia-inducible factor 1α (HIF-1α) mRNA by suppressing its recruitment to EIF4E⁶⁷. Following mitosis completion, phosphorylation of Ser1186 on EIF4G1 enhances its interaction with MNK1 (an EIF4E kinase), thereby promoting MNK1-mediated phosphorylation of Ser209 on EIF4E. This event ultimately reactivates translation initiation by enabling recruitment to mRNAs^68–70. Further, in brain injury induced by ischemia-reperfusion, global translation is broadly inhibited. However, increased phosphorylation of Ser1147 on EIF4G1 enables its competitive binding to EIF4E, maintaining a low level of translation to meet the synthesis of apoptosis-related proteins, thereby driving delayed neuronal death⁷¹. In this study, however, phosphoproteomic profiling identified that PPP6R3 deficiency does not affect EIF4EBP1 phosphorylation. Instead, our findings demonstrate that PPP6R3 deficiency destabilizes EIF3C and EIF4G1 proteins through enhancing the phosphorylation of Ser39 on EIF3C and Ser1217 on EIF4G1. This mechanism may parallel the proteasomal degradation of phosphorylated IκB (inhibitor of nuclear factor kappa B) via the IKKs-SCF ubiquitin ligase axis, wherein IKKs (IκB kinases) phosphorylate specific serine residues of the substrate IκB. The SCF (Skp1-CDC53-F-box) complex then conjugates ubiquitin to the phosphorylated serine, targeting the substrate for 26S proteasome degradation^72–75. While this hypothesis requires further experimental validation, our data conclusively establish that EIF3C and EIF4G1 are the substrates of PPP6R3/PP6 holoenzyme at Ser39 and Ser1217 sites, respectively. Elevated PPP6R3 level suppresses phosphorylation of these sites, stabilizing EIF3C and EIF4G1 as well as preserving ribosome assembly. This stabilization is critical for the transition of undifferentiated spermatogonia to differentiating spermatogonia, thereby maintaining normal spermatogenesis.

Finally, the RNA-dependent interactions between PPP6R3 and EIF3B, EIF3F, EIF3M, and EIF4E suggest that these associations may arise through competitive or cooperative binding to shared RNA elements, rather than direct protein-protein contact. These findings imply that PPP6R3 may regulate translation initiation through dual mechanisms: (1) direct interaction with core initiation factors (EIF3C, EIF4G1) to stabilize ribosome assembly, and (2) RNA-mediated cross-talk with accessory factors (EIF3B, EIF3F, EIF3M, and EIF4E) to modulate context-dependent translational efficiency.

Methods

Mice

We generated Ppp6r3^flox/flox mice on a C57BL/6 genetic background in collaboration with GemPharmatech Co., Ltd. Based on the expression profile of PPP6R3, which is low in undifferentiated spermatogonia and high in differentiating spermatogonia and spermatocytes, the role of PPP6R3 in spermatogonial differentiation (rather than stem cell function) was investigated using Stra8 GFP-cre mice, generously provided by Prof. Ming-Han Tong from the Center for Excellence in Molecular Cell Science, Chinese Academy of Sciences. To obtain Ppp6r3^flox/+ Stra8 cre mice, we crossed 8-week-old Stra8 GFP-cre males with 7-week-old Ppp6r3^flox/flox females. Subsequently, we mated 8-week-old Ppp6r3^flox/+ Stra8 cre males with 7-week-old Ppp6r3^flox/flox females to produce Ppp6r3^flox/flox Stra8 cre mice, referred to as Ppp6r3-cKO mice. Exon 4 of Ppp6r3 was selected for knockout. All mice were maintained in a specific-pathogen-free (SPF) facility under controlled conditions at 25 °C with a 12-hour light/dark cycle, with ad libitum access to food and water. No unexpected adverse events, such as infection or early death, were observed. Mice were euthanized by CO₂ asphyxiation or alternatively with sodium pentobarbital anesthesia overdose followed by cervical dislocation. All procedures adhered to the National Research Council Guidelines for the Care and Use of Laboratory Animals, and the study was approved by the Animal Ethics Committee of the School of Medicine, Shandong University. We have complied with all relevant ethical regulations for animal use.

Genotyping

We extracted genomic DNA from mouse tails and conducted PCR amplification using primers specific to both the Ppp6r3 mutant and wild-type alleles. PCR products were 363 bp for wild-type mice and 558 bp for Ppp6r3-cKO mice. The primers sequences used for genotyping are listed in Table S2.

Histological analysis

Testes were collected immediately after euthanasia and fixed in either 4% paraformaldehyde or Bouin’s solution (Sigma-Aldrich, HT10132). After dehydration, the tissues were embedded in paraffin and sectioned into 5 μm slices. Sections fixed in Bouin’s solution were subsequently stained with hematoxylin. Images were captured using an Olympus BX53 fluorescence microscope.

Immunostaining

We performed immunofluorescence on cultured cells as previously described⁷⁶. Briefly, the cells were fixed with 4% paraformaldehyde for 30 min, followed by treatment with 0.5% Triton X-100 for 20 min at 25 °C. This step was omitted when staining membrane proteins, such as GFRα1 and KIT. For immunostaining of testicular sections, we used a 1× boiling sodium citrate antigen retrieval buffer (pH 6.0) (Proteintech, PR30001) for 20 min. After blocking with 5% BSA in PBS for 1 h, both sections and cells were incubated with primary antibodies overnight at 4 °C, followed by secondary antibody incubation for 1 h at 25 °C. Samples were washed three times with PBS containing 0.05% Tween 20 (10 min each), Nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI). Images were captured using a confocal microscope (Andor Dragonfly spinning disc confocal microscope controlled by Fusion Software), and processed with Bitplane Imaris software (version 8.1).

For double-immunofluorescence staining of PPP6R3 and SOX9 on tissue sections, we used a Double-Fluorescence Immunohistochemical Mouse/Rabbit kit (pH 9.0) (ImmunoWay, RS0036) as both primary antibodies for PPP6R3 and SOX9 were anti-rabbit, following the manufacturer’s instructions. Antibody details are provided in Table S1.

Chromosome spread immunofluorescence analysis

We prepared chromosome spreads following previously described protocols^77,78. Briefly, testes were incubated in a hypotonic solution containing 30 mM Tris, 50 mM sucrose, 17 mM trisodium citrate, 5 mM EDTA, and 0.5 mM DTT for 30 to 40 min. Testes were then transferred into 100 mM sucrose, minced, and dropped onto slides pre-coated with a solution of 1% PFA and 0.15% Triton X-100. Once dried and washed with PBS, the slides were immunostained for SYCP1, SYCP3, and γH2AX.

RT-PCR and qRT-PCR

Total RNA was extracted from testes or KIT⁺ spermatogonia, isolated from the testis of wild-type and Ppp6r3-cKO mice, using TRIzol reagent according to the manufacturer’s protocol. The RNA was then reverse-transcribed into cDNA, followed by RT-PCR and qRT-PCR analyses. The primers sequences for RT-PCR and qRT-PCR are listed in Table S2.

Western blotting

Nucleoplasmic proteins were extracted from testes of wild-type mice using the Cytosolic and Nuclear Extraction kit for frozen/fresh tissues (Invent, NT-032), following the manufacturer’s instructions. Total protein lysates were obtained with RIPA buffer, and protein concentrations were measured using the BCA protein quantification kit. Equal amounts of protein were separated via SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked with 5% milk and incubated with primary antibodies overnight at 4 °C. After three 10-minute washes with TBS-T, secondary antibodies were applied for 1 h at 25 °C. Immunoblots were visualized using the Bio-Rad ChemiDoc MP Imaging System, and the band intensities were quantified using ImageJ software.

To assess phosphorylated and non-phosphorylated levels of translation initiation factors, we used a Phos-assay Acrylamide kit, according to the manufacturer’s protocol, and 5% BSA was used instead of 5% milk for blocking. Antibody details are provided in Table S1.

Purification of KIT⁺ spermatogonia

KIT⁺ spermatogonia were isolated and purified using the MojoSort^TM Mouse CD177 (c-Kit) Selection kit (Biolegend, 480146). Briefly, testes from post-natal day 9 (P9) wild-type mice were digested with 1 mg/mL collagenase IV and 0.05% trypsin. After passing the suspension through a 40-μm cell strainer to remove undigested material, cells were incubated sequentially with TruStain FcX^TM (anti-mouse CD16/32) and PE anti-mouse CD117 (c-Kit) antibodies. Following washes with MojoSort^TM Buffer, mouse anti-PE Nanobeads were used to enrich the KIT⁺ spermatogonia. The purity of the sorted cells was confirmed by immunostaining against KIT.

Isolation and culture of spermatogonial progenitor cells (SPCs)

We obtained single-cell suspensions from the testis of one wild-type mouse and one Ppp6r3-cKO mouse using a two-step enzymatic digestion process. We then purified THY1⁺ spermatogonia by incubating the cells with Anti-Mouse CD90.2 magnetic particles at 4°C for 30 min. These THY1⁺ spermatogonia were cultured on mitomycin C-inactivated mouse embryonic fibroblast cells in SSC medium⁷⁹, as previously reported, without insulin, in 12-well plates. We changed the culture medium every two days and subcultured the SPCs every 5 to 7 days at a ratio of 1:4. After five months of cell culture, the characteristics of SPCs were verified by RT-PCR and immunostaining.

CCK-8

To assess whether PPP6R3 deficiency impacts the viability of SPCs, we seeded approximately 5000 wild-type and Ppp6r3-cKO SPCs onto 96-well plates coated with a feeder layer. Following 7 days of culture, 10% (v/v) CCK-8 (Cell Counting Kit-8) reagent was added to the medium and incubated for 2 h. Absorbance was subsequently measured at 450 nm to quantify cell viability.

EdU Staining

To assess the impact of PPP6R3 deletion on the proliferative capacity of SPCs, we employed an EdU incorporation assay (MedChemExpress, HY-K1085). Given that SPCs grown on a feeder layer form dense clonal clusters that hinder accurate quantification of EdU-positive cells, we seeded approximately 5000 wild-type and Ppp6r3-cKO SPCs onto 96-well plates coated with Matrigel (1:80 dilution, Corning, 354277). After 7 days of culture, cells were fixed and stained according to the EdU kit protocol. The EdU positive rate was then quantified to evaluate cell proliferation.

Differentiation of SPCs in vitro

When the density of SPC clones (within 10 passages) reaches more than 80%, we digested both wild-type and Ppp6r3-cKO SPCs with 0.05% trypsin, then transferred these cells into 12-well plates coated with 0.2% gelatin for 1 h to remove feeder cells as thoroughly as possible. Approximately 10,000 SPCs were transferred to 24-well plates coated with Matrigel (1:80 dilution, Corning, 354277). After 24 h, we replaced the SSC medium with differentiation medium composed of Minimum Essential Medium Alpha, 2 mM GlutaMAX, 1× NEAA, 1× antibiotic-antimycotic, 60 μM putrescine, 60 ng/mL progesterone, 25 μg/mL insulin, 10 μM β-mercaptoethanol, 1 μM retinoic acid (RA), and 10% fetal bovine serum. The differentiation medium was refreshed daily.

Transcriptomics

We collected testis samples from P9 wild-type and Ppp6r3-cKO mice for transcriptomics profiling. We extracted total RNA using TRIzol reagent and enriched mRNA with Oligo(dT) magnetic beads. After generation of cDNA libraries, we performed Illumina sequencing. We aligned the paired-end clean reads to the mouse mm10 genome using Hisat2 (v2.0.5). Novogene Co., Ltd. (Beijing, China) provided the transcriptomics services.

Proteomics and phosphoproteomics

We extracted total protein from the testis of P9 wild-type and Ppp6r3-cKO mice, followed by digestion with trypsin and desalting. For proteomics, we separated a portion of the digested products and analyzed them using high-precision mass spectrometry. We utilized database retrieval software, including Proteome Discoverer, Mascot, Spectronaut, and MaxQuant, for data analysis. In the phosphoproteomics analysis, we enriched phosphopeptides using TiO₂ and subsequently eluted them. We identified and quantified phosphorylation sites through Mascot software after conducting liquid chromatography-tandem mass spectrometry (LC-MS/MS). The National Center for Protein Sciences (Beijing, China) provided the Proteomics and phosphoproteomics services.

Immunoprecipitation mass spectrometry (IP-MS)

We lysed purified KIT⁺ spermatogonia using pierce^TM IP lysis buffer (Thermo Scientific, 87787) supplemented with a protease inhibitor. We incubated the lysates with PPP6R3 and IgG antibodies overnight at 4 °C. To capture the protein-antibody complexes, we utilized protein A/G magnetic beads (Selleck, B23202) and eluted the bound proteins for mass spectrometry analysis. Nanjing JiangBei New Area Biopharmaceutical Public Service Platform Co., Ltd. (Nanjing, China) processed the IP-MS services.

Protein stability assays

To assess the stability changes of EIF3C and EIF4G1 in KIT⁺ spermatogonia following PPP6R3 deletion, we cultured purified KIT⁺ spermatogonia from testes of P9 wild-type and Ppp6r3-cKO mice in 12-well plates using SPC differentiation medium supplemented with 10 ng/mL cycloheximide for 0, 3, 6, and 9 h. We then collected the cells and evaluated the degradation rates of EIF3C and EIF4G1 by Western blotting.

RIP-qPCR

We measured the enrichment of EIF3C and EIF4G1 in relation to spermatogonial maintenance and differentiation-related mRNAs using a RIP kit (BersinBio, Bes5101) according to the manufacturer’s guidelines. After digesting the testis with polysome lysis buffer containing protease and RNase inhibitors, we removed DNA with DNase treatment. We incubated the lysates overnight at 4°C with antibodies against EIF3C, EIF4G1, and IgG. Then, we extracted the RNA bound to the target proteins. We performed qRT-PCR to quantify the enrichment of target mRNAs associated with EIF3C and EIF4G1, normalizing the results to 1% input. The primers sequences for RIP-qPCR are listed in Table S2.

Detection of target mRNA translation rate

We assessed the translation rates of mRNAs associated with spermatogonial maintenance and differentiation using a modified version of the “Targeted Profiling of RNA Translation” method described previously^36,37. This method employs a systematic protocol involving sequential steps: cycloheximide treatment of cultured cells, rapid sample collection, isolation of ribosome-protected mRNA footprints, primer-directed reverse transcription (RT) targeting translation initiation sites (TIS) that simultaneously modifies the properties of the resulting cDNA templates, and subsequent quantitative PCR (qPCR) analysis of these modified templates. In brief, wild-type and Ppp6r3-cKO SPCs were collected after 0, 6, and 12 h of differentiation. Cells were lysed using a buffer containing 20 mM Tris (pH 7.4), 150 mM NaCl, 5 mM MgCl₂, 1 mM DTT, 100 μg/mL cycloheximide, 1% Triton X-100, and 25 U/mL Turbo DNase. To digest RNA fragments outside ribosomes, we incubated the lysates with 100 U/μL RNase I at 22 °C for 45 min on a thermomixer at 1000 rpm. The digestion was stopped by adding an RNase inhibitor, followed by RNA extraction and reverse transcription. qPCR was then performed, and the translation rate of multiple spermatogonial maintenance and differentiation-related mRNAs was normalized to gapdh. Primers sequences used for this analysis are listed in Table S2.

Cloning and generation of EIF3C and EIF4G1 mutant variants

Mutant variants of EIF3C and EIF4G1 (EIF3C^S39A, EIF4G1^S1217A) were generated through standard overlapping PCR. These variants were cloned into the pcDNA3.1 plasmid, with a 3× FLAG tag attached at the C-terminus.

Overexpression of EIF3C and EIF4G1 mutant variants in SPCs

We overexpressed the EIF3C^S39A and EIF4G1^S1217A mutants, alongside a negative control (pcDNA3.1 plasmid), in Ppp6r3-cKO SPCs using the SP100 electroporation system (Celetrix, 11-0103). Ppp6r3-cKO SPCs were first digested with 0.05% trypsin and feeder cells were removed through differential plating. Three million Ppp6r3-cKO SPCs were resuspended in 100 μL electroporation buffer (equal parts buffer A and B) containing 10 μg plasmid. Electroporation was conducted in triplicate at 500 V for 20 ms. Following electroporation, cells were cultured on mitomycin C-inactivated mouse embryonic fibroblast cells with SPC medium for 6-8 days. Transfected cells were then transferred to a matrigel-coated 12-well plate, where 200 μg/mL neomycin was added to the SSC medium to select successfully transfected cells.

Statistics and Reproducibility

All quantitative data are presented as mean ± SD, based on at least three biological replicates. Statistical significance was determined using an unpaired two-tailed Student’s t-test. For transcriptomics, proteomics, phosphoproteomics, and IP-MS analyses, a fold change greater than 1.5 and p < 0.05 were considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

Q.F. conceived and designed the project, conducted the majority of experiments, collected and analyzed the data, and drafted the manuscript. B.L. performed the purification of KIT⁺ spermatogonia, carried out IP-MS, and cultured SPCs with assistance from Q.F. J.C. identified the abnormal meiosis process in spermatocytes of Ppp6r3-cKO mice. T.L. analyzed the transcriptomics, proteomics, and phosphoproteomics data. S.J. performed proteomics and phosphoproteomics sequencing. W.L., G.L., X.W., and J.B. provided constructive suggestions for data analysis. Z.J.C. and H.L. initiated and supervised the project and provided revisions to the manuscript.

Peer review

Peer review information

Communications Biology thanks Hiromitsu Tanaka, Fayçal Boussouar, and the other, anonymous, reviewer for their contribution to the peer review of this work. Primary Handling Editors: Martina Rauner and David Favero.

Data availability

All uncropped blot/gel images presented in the manuscript can be found in Fig. S7–S11. EIF3C^S39A and EIF4G1^S1217A mutant plasmids have been deposited in Addgene (Deposit Number: 86173). The source data for the graphs are provided in Supplementary Data. The transcriptomics data have been deposited in the NCBI GEO database (GSE278121). The mass spectrometry proteomics data of PPP6R3 based on KIT⁺ spermatogonia and the testis-based proteomics and phosphoproteomics raw data have been deposited to the ProteomeXchange Consortium via the iProX partner repository with the data identifiers PXD056272, PXD056273, and PXD056274, respectively. All other data are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s42003-025-08539-1.